AI Paper Atlas

Research map / Detection and segmentation / CNN features

Less is More: Focus Attention for Efficient DETR

Dehua Zheng, Wenhui Dong, Hailin Hu, Xinghao Chen, Yunhe Wang

ICCV 2023 · Detection and segmentation › CNN features · approach: Efficient DETR query design · 168 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace Sparse DETR token pruning with Focus-DETR on DINODINO · COCO val2017, keep ratio 0.3AP: 48.2 -> 50.4Improved
use Focus-DETR instead of full DINODINO · COCO val2017AP and compute: AP 50.9 -> 50.4; GFLOPs 279 -> 154; FPS 14.2 -> 20.0Mixed
apply Sparse DETR to DETR variants with learnable queriesDeformable DETR, DN-Deformable-DETR, DINO · COCO val2017AP: -1.9, -1.2, -2.7 APHurt
add supervision with improved label assignment to foreground token selector (ablation)Focus-DETR · COCO val2017, ResNet-50AP: 47.8 -> 48.8Improved
add top-down score modulation (ablation)Focus-DETR · COCO val2017, ResNet-50AP: 48.8 -> 49.2Improved
add cascade token selection (ablation)Focus-DETR · COCO val2017, ResNet-50AP: 49.2 -> 49.7Improved
add dual attention encoder (ablation)Focus-DETR · COCO val2017, ResNet-50AP: 49.7 -> 50.4Improved
use top-down instead of bottom-up score modulation (ablation)Focus-DETR · COCO val2017, ResNet-50AP: 50.2 -> 50.4Improved
add dual attention to models using Sparse DETR scoring without Focus-DETR scoring (ablation)Deformable DETR, DN-Deformable-DETR, DINO · COCO val2017AP: +0.3, 0.0, -0.1, +0.3 APNo clear effect
increase keep ratio from 0.3 to 0.5 (ablation)Focus-DETR · COCO val2017, ResNet-50AP: 50.4 -> 50.5No clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy